Machine Learning Techniques for Bird Sound Classification

Conservation of biodiversity is necessary for our existence. Birds are an integral part of biodiversity. Birds play an important role in controlling the functioning of ecosystem. Monitoring a region for presence of bird species is an important task required for conservation of biodiversity. Hence, monitoring avian behavior accurately and efficiently will be beneficial to the mankind. Acoustic communication in birds is rich and is the most important way by which they make their presence felt in the surroundings. Birds are difficult to be sighted, and therefore, their sounds are the most direct ways for the humans to detect them. In order to assess the biodiversity of a region, there is a great need for development of automatic analysis of avian sounds. This book discusses machine learning techniques used to analyze, classify and recognize bird species. Recognizing bird species, through their sounds, is a typical pattern recognition problem. Audio classification basically involves vital temporal, spectral and statistical features and classifiers. In this research, various feature extraction methods and classifiers are compared in order to develop a system with high accuracy.

Autorentext
Arti V. Bang received her Bachelor of Engineering from Marathwada University (India) in 1990, and Masters of Engineering from University of Pune (India) in 2000, and Doctorate from COEP, Pune, India in 2018. Dr. Arti worked in COEP, Pune, as a lecturer. Currently, she is working as Associate Professor in VIT, Pune.